The WSJ has published a excerpt of a speech from the late Michael Crichton. I would highly recommend it, as it is one of the few places I have seen a critical discussion of using model-generated data as a proxy for real data. My unwillingness to jump on the environmental bandwagon is largely grounded in this distinction.
And for anyone about to step up and defend those models, may I remind you of the many financial models that helped facilitate the current crisis. If someone can point out a fundamental difference between the two types of models that should make environmental models more trustworthy than financial models, I'd welcome the enlightening.